19 Jul
2009
19 Jul
'09
9:26 p.m.
I have an ensoniq audiopci card that I have been using successfully at
48000 hz for a while now. I have just started working on a project
with prerecorded material at 44100 hz, so I reconfigured jack in order
to avoid needless sample rate conversion. Somehow jack thinks that my
card is running at 44099 hz, so it is resampling the source material
anyway.
Why would my card be running at 44099 hz, is this a software or
hardware problem? I am going to just edit the wav headers to say 44099
so they agree with the engine sample rate for now, but would be
interested to know what the source of this weirdness would be.
19 Jul
19 Jul
10 p.m.
On Sun, Jul 19, 2009 at 7:26 PM, Justin Smith<noisesmith(a)gmail.com> wrote:
I have an ensoniq audiopci card that I have been using
successfully at
48000 hz for a while now. I have just started working on a project
with prerecorded material at 44100 hz, so I reconfigured jack in order
to avoid needless sample rate conversion. Somehow jack thinks that my
card is running at 44099 hz, so it is resampling the source material
anyway.
Why would my card be running at 44099 hz, is this a software or
hardware problem? I am going to just edit the wav headers to say 44099
so they agree with the engine sample rate for now, but would be
interested to know what the source of this weirdness would be.
its reported by the ALSA driver, which in turn is written based on
information available about the h/w. its debatable what ALSA should
report in these situations, but its a driver-level issue, and not
something in JACK's domain.
JACK does NOT resample audio, ever. neither does ALSA if you use the
hw:N device(s).
10:03 p.m.
On Sun, Jul 19, 2009 at 5:00 PM, Paul Davis<paul(a)linuxaudiosystems.com> wrote:
On Sun, Jul 19, 2009 at 7:26 PM, Justin
Smith<noisesmith(a)gmail.com> wrote:
Sorry, I was conflating jack and ardour, I meant that ardour was
wanting to resample.
I used sox to copy the audio files to .raw, and then converted the
.raw files to .wav at 44099 sampling rate. I had a suspicion it may
have been the alsa backend, thanks for the info.
I have an ensoniq audiopci card that I have been
using successfully at
48000 hz for a while now. I have just started working on a project
with prerecorded material at 44100 hz, so I reconfigured jack in order
to avoid needless sample rate conversion. Somehow jack thinks that my
card is running at 44099 hz, so it is resampling the source material
anyway.
Why would my card be running at 44099 hz, is this a software or
hardware problem? I am going to just edit the wav headers to say 44099
so they agree with the engine sample rate for now, but would be
interested to know what the source of this weirdness would be.
its reported by the ALSA driver, which in turn is written based on
information available about the h/w. its debatable what ALSA should
report in these situations, but its a driver-level issue, and not
something in JACK's domain.
JACK does NOT resample audio, ever. neither does ALSA if you use the
hw:N device(s).
10:55 p.m.
On Sun, Jul 19, 2009 at 8:03 PM, Justin Smith<noisesmith(a)gmail.com> wrote:
Sorry, I was conflating jack and ardour, I meant that
ardour was
wanting to resample.
I used sox to copy the audio files to .raw, and then converted the
.raw files to .wav at 44099 sampling rate. I had a suspicion it may
have been the alsa backend, thanks for the info.
damn. that is irritating as all hell... i'll have to ponder what to do
about that. i can think of a relatively simple fix in ardour.
20 Jul
20 Jul
9 a.m.
On Monday 20 July 2009 02:26:44 Justin Smith wrote:
I have an ensoniq audiopci card that I have been using
successfully at
48000 hz for a while now. I have just started working on a project
with prerecorded material at 44100 hz, so I reconfigured jack in order
to avoid needless sample rate conversion. Somehow jack thinks that my
card is running at 44099 hz, so it is resampling the source material
anyway.
Why would my card be running at 44099 hz, is this a software or
hardware problem? I am going to just edit the wav headers to say 44099
so they agree with the engine sample rate for now, but would be
interested to know what the source of this weirdness would be.
I filed this as a bug on the alsa driver more than once. The problem is in the
Ensonic PCI card. Its firmware has the round-off error (obviously) and reports
44099. (This is not a pro-audio card. Unfortunately, my old dman has no alsa
support.)
This is a royal pain in the you-know-what! Some apps will simply not start up.
I can get Tapestry to run by specifying 44099 on its command line! Others,
like yours, may want to resample. Pain.
I believe that alsa's drivers should not report back absurd data. There is no
such animal as 44099. It should be changed to 44100.
10:57 a.m.
David,
I believe that alsa's drivers should not report
back absurd data. There is no
such animal as 44099.
Of course there is, and some cards can generate arbitrary word clocks
internally. It es even necessary for some applications, e.g. proper
synchronization over the net.
(In your case, this may be a bug nontheless :)
Flo
--
Machines can do the work, so people have time to think.
public key DA43FEF4 x-hkp://wwwkeys.eu.pgp.net
5759
days inactive
5760
days old
linux-audio-user@lists.linuxaudio.org
5 comments
4 participants
participants (4)
-
David Baron -
Florian Faber -
Justin Smith -
Paul Davis